Suppression of human pancreatic tumor growth in vivo by sulindac and a novel analog of parthenolide with improved bioavailability

Abstract

4949

Background: Cancer of the exocrine pancreas is the fourth leading cause of cancer-related deaths in the United States. Novel chemotherapeutic and chemopreventive options are needed for this disease. Parthenolide, an inhibitor of nuclear factor-kappaB (NF-κB), and sulindac, a cyclooxygenase (COX) inhibitor, show complementary antiproliferative effects on pancreatic cancer cells in vitro. An analog of parthenolide, (LC-1), has recently been synthesized that is more bioavailable. The aim of the present study was to evaluate the efficacy of LC-1 and/or sulindac in a xenograft model of pancreatic cancer in vivo. Methods: Proliferation (MTT) and NF-κB DNA binding assays were performed in the human pancreatic cancer cell line PaCa-2 to determine whether parthenolide and LC-1 had equivalent activities in vitro. To determine the dose of sulindac for in vivo studies, the effect of sulindac on prostaglandin E2 (PGE₂) levels was determined ex vivo by ELISA in xenografted BxPC-3 pancreatic tumors 20 hours following a single oral administration of sulindac. For in vivo studies, nude mice were injected with BxPC-3 (COX-2 positive) or PaCa-2 (COX-2 negative) human pancreatic cancer cells. Drug treatment was initiated on the day of injection. Mice were randomly divided into six treatment groups that included placebo, low/high dose LC-1 (10 or 20 mg/kg, 40 mg/kg), low/high dose sulindac (20 and 60mg/kg), and combination of low dose LC-1/sulindac. Tumor size was measured twice weekly. Mean tumor volumes for each treatment group were calculated. Results: Parthenolide and LC-1 demonstrated equivalent antiproliferative effects and suppression of NF-κB DNA binding activity in PaCa-2 cells; in addition, similar cooperative repression of cell growth was observed in combination with sulindac in vitro. In ex vivo studies, 20 hours after treatment with low dose (20mg/kg) and high dose (60mg/kg) sulindac, PGE₂ levels (2.1 +/− 1.4 and 1.33 +/− 0.5 pg PGE₂/ug total protein, respectively) were decreased relative to placebo (3.86 +/− 0.5 pg PGE₂/ug). The in vivo trial showed that the combination of 20mg/kg LC-1 and 20mg/kg sulindac suppressed the growth of BxPC-3 (COX-2 positive) xenografted tumors relative to placebo (P<0.05, t-test). Sulindac (60mg/kg) alone at earlier timepoints showed significant suppression of growth (P<0.05, t-test). LC-1 alone did not have any significant effect. In the PaCa-2 (COX-2 negative) xenografted tumors, the LC-1 and combination groups exhibited no significant change in growth. Interestingly, sulindac (60mg/kg) alone at the higher dose inhibited tumor growth relative to placebo (P<0.05, t-test). Conclusions: The parthenolide analog LC-1 and sulindac in combination inhibit tumor growth effectively in a COX-2 positive pancreatic tumor. Sulindac alone at a higher dose suppresses growth of a COX-2 negative pancreatic tumor, likely through a COX-independent mechanism.